{
  "schema_version": "paper_public_manifest_v1",
  "paper_id": "mathcaln_0_foundation_towards_the_age_of_tactile_intelligence_2026_09",
  "slug": "mathcaln_0_foundation_towards_the_age_of_tactile_intelligence",
  "title": "N0-Foundation:开启触觉新时代",
  "authors": [],
  "source": {
    "arxiv_id": "2608.29601",
    "pdf_url": "https://arxiv.org/pdf/2608.29601",
    "project_url": "",
    "github_url": "",
    "huggingface_url": "",
    "original_source": "https://arxiv.org/abs/2608.29601"
  },
  "site": {
    "post_url": "/posts/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence",
    "canonical_url": "https://haiguangboy.com/posts/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence",
    "cover_image": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/cover.webp"
  },
  "taxonomy": {
    "domain": "embodied_ai",
    "track": "action_generation",
    "tasks": [
      "embodied_ai",
      "action_generation",
      "robotics",
      "state_prediction",
      "具身智能",
      "触觉智能",
      "力场表示"
    ],
    "related_topics": [
      {
        "paper_id": "prism_precision_and_contact_rich_real_world_industrial_skill_dataset_with_multim_2026_08",
        "title": "PRISM:规模能否替代显式建模?",
        "url": "https://haiguangboy.com/posts/prism_precision_and_contact_rich_real_world_industrial_skill_dataset_with_multim",
        "relation": "contrast",
        "summary": "PRISM:规模能否替代显式建模?",
        "strength": "strong"
      },
      {
        "paper_id": "tacwam_anchor_guided_world_action_model_with_mechanics_aware_tactile_prediction_2026_08",
        "title": "触觉预测只该用来训练，不该给动作看",
        "url": "https://haiguangboy.com/posts/tacwam",
        "relation": "contrast",
        "summary": "作者判断:现有策略在精细接触密集任务上的瓶颈,根本原因是纯视觉+本体感知既暴露不出接触状态、也提供不了动作执行中途纠错所需的信号——不是策略容量不够,是感知模态本身缺了这一环 contradicts 核心原则:未来触觉只做训练期监督信号,不进入部署期动作输入",
        "strength": "strong"
      },
      {
        "paper_id": "jepa_wam_learning_vision_language_action_policies_with_joint_embedding_world_modeling_2026_08",
        "title": "JEPA-WAM：创新方法让世界模型更懂状态变化",
        "url": "https://haiguangboy.com/posts/jepa-wam",
        "relation": "same_track",
        "summary": "触觉集成方式消融(固定π0.5骨干):不给触觉<原始触觉图像拼接<原始触觉图像喂给动作专家<NeoForce力场喂给动作专家,平均渐进得分从38.1%升到47.5% validates 核心主张:转移监督必须直接塑造生成动作的那个主干,外挂和当上下文都不行",
        "strength": "strong"
      },
      {
        "paper_id": "skild_blog_introducing_s1_in_context_learning_20260817_2026_08",
        "title": "S1:机器人的GPT-3涌现时刻",
        "url": "https://haiguangboy.com/posts/skild_blog_introducing_s1_in_context_learning_20260817",
        "relation": "same_track",
        "summary": "S1:机器人的GPT-3涌现时刻",
        "strength": "strong"
      },
      {
        "paper_id": "t_rex_tactile_reactive_dexterous_manipulation_2026_07",
        "title": "T-Rex：触觉为何要单独建模",
        "url": "https://haiguangboy.com/posts/t_rex_tactile_reactive_dexterous_manipulation",
        "relation": "same_track",
        "summary": "T-Rex：触觉为何要单独建模",
        "strength": "strong"
      },
      {
        "paper_id": "pose_aware_modeling_to_mitigate_pose_related_artifacts_in_tactile_gloves_2026_07",
        "title": "手套触觉读数里混着手的姿态",
        "url": "https://haiguangboy.com/posts/tactile-glove-pose-artifacts",
        "relation": "same_track",
        "summary": "触觉传感器是视觉式(camera-based visuo-tactile):弹性感应层受力形变,内置相机拍下纹理位移,再学一个逆映射把图像还原成力场,而不是直接把原始图像当触觉表示用 validates 又一次'直接拼接会更差':这次的解释是模型学错了映射方向",
        "strength": "strong"
      }
    ]
  },
  "analyst_take": {
    "type": "author_opinion",
    "text": "跟TacWAM对比最尖锐:TacWAM说未来触觉只能做训练期监督、不能进部署期动作输入,这篇消融却显示把力场喂给部署时的动作专家收益最大——两边给出相反答案;不过TacWAM那条\"触觉存在结构性时间歧义、单帧不够\"的诊断,这篇的力场表示专门在片段内聚合时序上下文来应对,算是间接认领了同一个问题。跟T-Rex是同一路线:触觉需要专门的异步处理、机械拼接无效,这篇没把原始触觉图像直接拼进策略,是同一判断的又一次独立印证;T-Rex真机12项任务成功率65%、比最强基线高出30个百分点,这篇量级更大,10项任务、3万+小时数据。跟N0-VTLA是同一脉络延续:两篇用同一个NeoReal基准,N0-VTLA当时47.2%对π0.5的29.4%,这篇扩到10任务、把π0.5拉到26.5%——但也继承了N0-VTLA同样的局限:评测主要在自建基准上做,强依赖自研传感器和自建数据管线。"
  },
  "ruling": {
    "importance_score": 3.0,
    "one_sentence": "N0-Foundation:开启触觉新时代"
  },
  "asset_base_url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence",
  "assets": [
    {
      "type": "pdf_screenshot",
      "object_key": "papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/page_01.webp",
      "content_type": "image/webp",
      "upload_status": "uploaded",
      "bucket": "paper-assets",
      "url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/page_01.webp",
      "role": "paper_first_page",
      "size_bytes": 194188
    },
    {
      "type": "pdf_screenshot",
      "object_key": "papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/key_figure.webp",
      "content_type": "image/webp",
      "upload_status": "uploaded",
      "bucket": "paper-assets",
      "url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/key_figure.webp",
      "role": "method_figure",
      "size_bytes": 182836
    },
    {
      "type": "cover_image",
      "object_key": "papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/cover.webp",
      "content_type": "image/webp",
      "upload_status": "uploaded",
      "bucket": "paper-assets",
      "url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/cover.webp",
      "role": "post_cover",
      "size_bytes": 92210
    },
    {
      "type": "public_brief",
      "object_key": "papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/public_brief.md",
      "content_type": "text/markdown; charset=utf-8",
      "upload_status": "uploaded",
      "bucket": "paper-assets",
      "url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/public_brief.md",
      "role": "public_brief",
      "size_bytes": 4999
    },
    {
      "type": "public_manifest",
      "object_key": "papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/public_manifest.json",
      "content_type": "application/json; charset=utf-8",
      "upload_status": "uploaded",
      "bucket": "paper-assets",
      "url": "https://static.haiguangboy.com/papers/mathcaln_0_foundation_towards_the_age_of_tactile_intelligence/public_manifest.json",
      "role": "public_manifest",
      "size_bytes": 7985
    }
  ],
  "published_at": "2026-09-04T13:04:44+08:00",
  "created_at": "2026-09-04T13:04:44+08:00",
  "updated_at": "2026-09-04T13:04:47+08:00"
}